Impact of Collective Electrostatic Effects on Charge Transport through

Aug 26, 2014 - Institut für Physik, Humboldt-Universität zu Berlin, Newtonstrasse 15, 12489 Berlin, ... of charge-transport properties with the size...
0 downloads 0 Views 7MB Size
Article pubs.acs.org/JPCC

Impact of Collective Electrostatic Effects on Charge Transport through Molecular Monolayers Veronika Obersteiner,† David A. Egger,*,†,‡ Georg Heimel,§ and Egbert Zojer*,† †

Institute of Solid State Physics, NAWI Graz, Graz University of Technology, Petersgasse 16, 8010 Graz, Austria Department of Materials and Interfaces, Weizmann Institute of Science, Rehovoth 76100, Israel § Institut für Physik, Humboldt-Universität zu Berlin, Newtonstrasse 15, 12489 Berlin, Germany ‡

S Supporting Information *

ABSTRACT: A microscopic understanding of charge transport through moleculebased systems is essential for advancing the field of molecular electronics. In the present paper we highlight fundamental differences between devices with an individual molecule or a homogeneous monolayer as the active element. These differences arise from collective electrostatic effects that govern the electronic level alignment in monolayer-based devices, but are absent in the case of individual molecules. Employing density functional theory in conjunction with a Green’s function approach, we show that depending on the chemical nature of the employed molecules collective electrostatic effects can result in either a significant increase or a significant decrease of the current per molecule as a function of the packing density, in certain cases even resulting in a change in the transport polarity. Understanding the underlying principles also allows for designing molecules in which dipolar effects cancel, and thus, the transport characteristics of individual molecules and monolayers remain similar.

1. INTRODUCTION The ultimate miniaturization of electrical components triggered paramount scientific interest in understanding charge transport on an atomistic length scale and led to what is known today as “molecular electronics”.1−3 The collective (or cooperative) behavior of individual molecules condensed into a molecular assembly is particularly important, as it determines the “scaling” of charge-transport properties with the size of a molecular junction.4 Experimentally, these scaling properties have been examined in several previous studies.5 For example, Kushmerick et al. observed the conductance to scale directly with the number of molecules present in a self-assembled monolayer (SAM) junction.6 Selzer et al., however, found that the scaling of conductance strongly depends on the applied bias and that the per-molecule current can strongly decrease due to environment effects.7 Similarly, also in theoretical studies,8−17 it was reported that, in going from one molecule to a few and further to many molecules, collective effects can “help or hamper” 15electronic transport. For example, direct molecular (π−π) and substratemediated coupling13,15 or quantum-interference phenomena18 were shown to generally increase the conductance per molecule. On the other hand, incoherent scattering can strongly decrease the per-molecule conductance in a SAM,16 and so can differences in the broadening of spectral properties when going from single molecules to monolayers.12,13 These previous findings clearly suggest that the evolution of chargetransport properties with the number of molecules is strongly system dependent. © XXXX American Chemical Society

While the above-mentioned effects originate mostly from quantum-mechanical interactions, in this study, the impact of collective electrostatic effects on charge-transport properties of molecular assemblies is systematically investigated. These arise from the peculiar way in which assemblies of dipoles influence the electrostatic potential-energy landscape (for recent reviews see refs 19−22). Such dipoles are present in the vast majority of molecular and monolayer junctions due to polar docking groups and/or interfacial charge rearrangements arising from the molecule−electrode bond formation. Previously, some of us have shown that the collective electrostatic interaction of dipolar units allows for switching the transport polarity in monolayer junctions.23 Here, we will exploit such intramolecular polar groups primarily to clearly illustrate the impact of collective electrostatic effects on charge-transport properties. We find that the detailed chemical structure of the molecules forming the junction determines whether the per-molecule current increases or decreases when the molecular packing density changes. We also show that, by designing molecules such that internal dipoles compensate bonding-related dipoles, it becomes possible to realize systems in which the current per molecule becomes largely packing independent. This could, for example, be exploited for reducing fluctuations arising from structural imperfections in monolayer devices caused by lateral variations of the local packing density. On the basis of wellReceived: August 22, 2014 Revised: August 26, 2014

A

dx.doi.org/10.1021/jp5084955 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

detailed description including depictions of all unit cells is contained in the Supporting Information. As a prototypical example of molecular and monolayer junctions, we studied “Tour-wire” 24 based systems, i.e., 1,2bis(2-phenylethynyl)benzene bonded to the Au electrodes with thiolate groups (TW molecule; see Figure 1b). The methylene bridges (−CH2−) are used to electronically decouple the πconjugated backbone from the metal to avoid substrate-induced effects that are not the focus of our present study. To investigate the impact of polar groups built into the molecular backbone,25 we followed the strategy from ref 23 and symmetrically replaced the two outermost phenyl rings with pyrimidine units, resulting in the Nout and the isomeric Nin molecules (see Figure 1b). Importantly, this procedure hardly affects molecular quantities; i.e., the frontier orbitals of Nout and Nin are delocalized π-states (see Supporting Information), and their ionization potential and electron affinity are essentially equal. However, the differently oriented C−N bonds lead to differently oriented “local” dipoles in the peripheries of the Nout and Nin molecules,23 indicated by black arrows in Figure 1b. Furthermore, all systems exhibit an additional local dipole, namely, through the polar C−S−Au bonds (see Figure 1b), comprising also the polarization response in the substrate.26 The collective effects due to these dipoles are the main focus of our study and will be explored in detail below. Note that the term “local dipole” is not used in a rigorous sense (which would require local charge neutrality), but rather to specify a local charge distribution that affects the electrostatic potential similarly to a (point) dipole. Overall, for symmetry reasons all three considered molecules do not possess a net dipole moment along the long molecular axis. Because our goal was to isolate the impact of collective electrostatics on charge transport, we chose to suppress several other effects usually superimposed in an actual junction. Therefore, we deliberately did not account for specifics in the junction geometry,27 disorder, or molecular orientation (varying tilt28 and azimuthal angles). Also the gap reduction due to polarization effects in the films is not considered.29−31 To nevertheless allow for a realistic structural setup of the studied systems, we fully optimized the SAM structures including the innermost gold layers and the length of the junction in the z direction (see Supporting Information for further details). To determine the electronic structure of the systems and optimize their geometry, we performed band-structure calculations within the framework of density-functional theory (DFT) using the VASP32 code and the Perdew−Burke− Enzerhof (PBE)33 exchange-correlation functional. For computing the charge-transport properties of the metal−molecule− metal junctions, we followed a three-step procedure: First, we used a locally modified version of the periodic electronicstructure code SIESTA34 for extracting the Hamilton and overlap matrix of the central region (comprising the abovedescribed unit cells, i.e., six layers of gold and the molecules at different packing densities). We applied a double-ζ polarized basis set. To achieve compatibility of the packing-dependent trends with the VASP results, we reduced the “PAO energy shift” to 0.001 Ry (see the Supporting Information for more information). Then we computed the self-energies of the leads in the recursive Green’s function formalism35 using a 2 × 2 × √6 unit cell for bulk gold, which was suitably increased at lower packing densities. Finally, the zero-bias transmission function, T(E), was calculated to obtain the current−voltage

established concepts of electrostatics and molecule−metal interfaces, our study reveals that transport measurements may or may not depend on the number of molecules present in the junction.

2. SYSTEM SETUP AND THEORETICAL METHODS In our calculations, we considered metal−molecule−metal junctions with varying lateral molecular packing densities, ⊖ (see Figure 1a). Employing periodic boundary conditions, we

Figure 1. (a) Structure of the densely packed (⊖ = 1; all molecules) junction and the model system for the single-molecule situation (⊖ = 1/16; dark molecule only). The unit cell for ⊖ = 1 is indicated by the blue box, while for ⊖ = 1/16 the whole displayed structure (excluding the shaded molecules) represents the unit cell used in the calculations. (b) Chemical structures of the investigated molecules TW, Nout, and Nin, including arrows indicating the relevant local dipoles present in the systems.

represented the leads present on each “side” of the junction by three layers of Au(111) (i.e., there are six layers of Au between periodic replicas of the organic monolayers). For the molecular part of the junction, we started from a densely packed (⊖ = 1) self-assembled monolayer (SAM) with one molecule per (2 × 2) Au(111) surface unit-cell. The molecular packing density was then gradually reduced by replicating the unit cell laterally in x and y direction, removing all but one molecule. In this way, we generated molecular packing densities ⊖ = 1/2, 1/4, 1/8, and 1/16. The latter, which corresponds to one molecule per 8 × 8 surface unit-cell, is a very sparse monolayer, which we here consider as a model for a single-molecule junction. This is justified by the electrostatic interactions between neighboring molecules becoming very small in the case of ⊖ = 1/16, as will be shown below. The procedure of varying the molecular packing density is schematically indicated in Figure 1a; a more B

dx.doi.org/10.1021/jp5084955 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

Figure 2. (a) Calculated current−voltage characteristics of the TW, Nout, and Nin systems for a full monolayer with a packing density ⊖ = 1 and the model for a single-molecule junction with ⊖ = 1/16. Thick solid lines denote total currents, whereas thin dashed lines denote the dominant types of current arising from electron (n-type) or hole (p-type) transport; see text for details. (b) Calculated transmission functions of the TW, Nout, and Nin systems for different packing densities ⊖. The Fermi level EF is used as the energy reference. The area below the curves for the highest and lowest considered packing densities is shaded.

(I−V) characteristics employing the Landauer−Büttiker36,37 approach as I (V ) =

2e h

∫ T(E)[f (E−μleft ) − f (E−μright )] dE

where the orientation of the polar C−N bonds is reversed, the current per molecule is almost independent of the molecular packing density. We have previously shown for densely packed Nin and Nout layers23 that local intramolecular dipoles can influence the polarity of charge transport through molecular monolayers. This determines, for example, the sign of the Seebeck coefficient42−44 and, thus, the thermoelectric properties of molecular or monolayer junctions. Therefore, it is interesting to investigate the charge-transport polarity for the different junctions also as a function of the packing density. To achieve this, we calculated the current through occupied and unoccupied states separately (by setting the remainder of T(E) to zero and re-evaluating eq 1), with the results shown as dashed lines in Figure 2a. At the highest packing density, the transport through TW and Nin SAMs is essentially p-type, while for the Nout SAM we find a more pronounced n-type current. When the packing density is reduced, again an intriguing system-dependent effect occurs: For TW and Nin, the chargetransport polarity observed at the highest packing density (ptype) prevails; for Nout, however, the polarity changes such that for the isolated molecule in the chosen configuration p-type and n-type conduction become very similar. 3.2. Evolution of the Transmission Functions with Molecular Packing Density. As current−voltage characteristics are directly affected by the energetic alignment of molecular transport channels relative to the Fermi level, EF, the above-discussed trends can be directly traced back to the transmission functions shown in Figure 2b. For TW and Nout, a reduction in the packing density results in a continuous shift of the transmission functions to more positive energies with respect to EF. Thus, in both cases the occupied channels approach EF, while the unoccupied channels retreat from it,

(1)

Here, f(x) is the Fermi−Dirac occupation function at 300 K and μleft/right = EF ± (eV/2), with e the elementary charge, EF the Fermi energy, and V the voltage. Because we rely on zerobias transmission functions, polarization effects due to an applied voltage are not included in our treatment. Nevertheless, qualitatively reliable results can be expected for current−voltage characteristics plotted in voltage regions far from resonances.38 OVITO,39 XCrySDen,40 and Mayavi41 were used for graphical visualization. For full details on the applied computational methodology and numerical parameters used in our calculations, see the Supporting Information.

3. RESULTS AND DISCUSSION 3.1. Charge-Transport Properties of Monolayers with Different Packing Densities. The resulting current−voltage (I−V) curves for TW, Nout, and Nin molecules at the highest (⊖ = 1) and lowest (⊖ = 1/16) considered packing densities are shown in Figure 2a (thick, solid lines). At ⊖ = 1, the transport characteristics of the three systems display certain differences: For example, for a per-molecule current of 0.5 μA to flow, voltages of 1.4, 1.0, and 1.2 V have to be applied for TW, Nout, and Nin, respectively. What is more surprising, however, are the distinctly different shifts of the chargetransport characteristics that are observed when the singlemolecule (i.e., low packing density) limit is approached: Reducing the molecular packing density in the junctions significantly increases the current per molecule for TW, whereas for Nout a marked decrease is observed. For Nin, C

dx.doi.org/10.1021/jp5084955 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

Otherwise, only relatively weak variations in the energy are visible, especially further away from the nuclei. In the Nout and Nin SAMs (see Figures 4 b,c) the situation changes dramatically compared to that in the TW monolayer: An additional, rather large potential energy gradient appears in the region of polar C−N bonds that strongly shifts the overall energy landscape for the molecules to more negative energies (Nout; blue shading in the central region of Figure 4b) and more positive energies (Nin; red shading in the central region of Figure 4c). These are “global” shifts, meaning that they affect the whole scattering region between the leads. Naturally, as such they also shift the electronic states in the monolayer relative to the Fermi level and, therefore, also the corresponding conducting channels in the transmission functions. We also find that these shifts are superimposed by local (i.e., molecular) perturbations of the potential energy, especially close to the N atoms, and in the case of the Nin SAM a weaker energy gradient along the carbon−carbon triple bonds that to some extent compensates the shift due to the pyrimidines. We attribute the latter, which is clearly a secondary effect, to a pyrimidine-induced polarization of that molecular region. The discussed global energy variations are fully consistent with the plane-averaged potentials (cf. Figure 3). When the molecular packing density is reduced, the potential energy shifts caused by the superposition of the fields of the dipoles arranged in parallel decrease continuously.45 This decrease depends sublinearly on the packing density owing to a smaller depolarization of C−N and C−S−Au dipoles at lower packing densities.46−48 Nevertheless, already for ⊖ = 1/16 (i.e., the lowest packing density considered here) only local potential energy variations which reflect the different natures of the involved chemical elements prevail. The “global” gradients in the direction perpendicular to the electrode surfaces have disappeared (see Figure 4 d−f). This clearly shows that, at low packing density, i.e., close to the single-molecule junction regime, the properties of the individual molecules dominate. In systems with interacting molecules, however, the collective electrostatic effects of superimposed dipole fields can significantly shift the electronic states in energy. These effects then determine the packing-dependent charge transport: At ⊖ = 1, the energy shift due to the C−S−Au dipole that occurs for all studied systems similarly adds to the effect of the pyrimidine dipole in the case of Nout. At lower packing densities, these shifts disappear, which explains the strong packing-density dependence of the transmission functions of TW and Nout. For Nin, however, at ⊖ = 1, the impact of the reversed pyrimidine dipole cancels that of the C−S−Au bond, giving rise to its packing-independent charge-transport properties. This shows that collective electrostatic effects can be effectively “switched off”.

with the effect significantly more pronounced in the Nout case (+0.3 eV vs +0.6 eV). As TW conducts via occupied channels, reducing the packing density reduces the tunnel barrier and consequently leads to an increase in the per-molecule current (see Figure 2a). For Nout, however, the current at ⊖ = 1 is dominated by electrons flowing through unoccupied states (see Figure 2a). Therefore, the energy shift of the transmission function causes a decrease in the current per molecule when the packing density is reduced. This eventually leads to the abovedescribed change in transport polarity at ⊖ = 1/16 (where for Nout the energetic distances of the occupied and unoccupied transport channels from EF become very similar). In the case of Nin, changes in the transmission function are comparably small, which is why the current and polarity are essentially independent of the molecular packing density. 3.3. Impact of Local Dipoles on the Electrostatic Potential Energy. Figure 3 shows the plane-averaged

Figure 3. Calculated plane-averaged electron electrostatic energy for the full packing density, ⊖ = 1, of TW (solid black line), Nout (dashed light-blue line), and Nin (dotted red line). The Fermi energy is set to zero.

electrostatic energy of an electron along the transport direction for all three systems at ⊖ = 1. Interestingly, the electrostatic energy in the region of the molecular backbone is quite different in the three systems, which is highly relevant for the charge-transport properties because this energetically shifts the delocalized orbitals (see the Supporting Information) that serve as transport channels. These distinct qualitative differences in the local electrostatics of the three considered systems can be understood considering the different orientations of the local dipoles, as schematically indicated by the black arrows in Figure 1b. As has been discussed extensively in the literature,19−22 a 2D periodic assembly of molecules with polar groups results in an energy shift left and right of the plane in which the dipoles are assembled. Interestingly, for point dipoles one can show that the decay length of the associated electric field is nearly an order of magnitude shorter than the dipole−dipole distance.19 To visualize the resulting abrupt changes in the electrostatic potential for the studied systems and analyze the effect of the local dipoles, we also plotted the electron electrostatic energy in a plane parallel to the central benzene rings for the densely packed SAMs, ⊖ = 1; see Figure 4 a−c. In the TW monolayer, the most prominent energy shift occurs in the region of the bond between the Au surface and the methyl thiolate bond (C−S−Au dipole). As the C−S−Au dipoles point toward the center of the junction, this effect shifts all transport channels in the monolayer to more negative energies with respect to EF.

4. CONCLUSIONS To summarize, we studied the impact of collective electrostatic effects on the charge-transport properties of molecular systems. Our results clearly demonstrate that these effects can change the transport characteristics of molecular ensembles even qualitatively as they strongly affect the total current as well as its polarity. Most importantly, we have shown that collective effects emerge naturally from the bonding of the molecules to (metallic) leads (i.e., they occur in virtually all monolayer junctions) and that they are strongly affected by the presence of polar bonds. Depending on the magnitude, orientation, and location of the associated dipoles, collective electrostatic effects D

dx.doi.org/10.1021/jp5084955 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

Figure 4. Contour plot of the electron electrostatic energy for the full packing density, ⊖ = 1 (a−c), and lowest packing density, ⊖ = 1/16 (d−f), of TW, Nin, and Nout in a plane parallel to the central benzene ring, approximately 1.6 Å away from that ring to avoid (i) extensive variations in the potential near the nuclei and (ii) artifacts caused by the bending of the molecules (see the Supporting Information). The choice of the plane in which the potential is plotted is also the reason for the potential maxima and minima halfway between the molecular backbones at highest packing densities. These originate from potential variations caused by the (bent) molecules in the neighboring unit cells. The potential energy is given relative to the respective Fermi level. Isolines are drawn every 0.2 eV; no isolines are drawn for electrostatic energies below 2.4 eV.

tions, and additional data. This material is available free of charge via the Internet at http://pubs.acs.org.

can significantly increase or decrease the current per molecule. They can be significant, shifting transmission maxima sometimes by more than 1 eV, and are, thus, larger than many of the otherwise discussed collective effects in molecular junctions. By appropriate molecular design, these effects can also be avoided. As collective electrostatic effects are operational only in (densely) packed monolayers but absent in individual molecules, the aforementioned trends directly map onto packing-dependent molecular conductivities. Consequently, the phenomena discussed here can play an important role in many of the relevant charge-transport scenarios and might help explain some of the seemingly conflicting observations of cooperative charge-transport behavior in molecular and monolayer systems.





AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interests.



ACKNOWLEDGMENTS We thank Leeor Kronik for stimulating discussions. Financial support by the Austrian Science Fund (FWF) (Grants P24666N20 and I937-N19) as well as by the Deutsche Forschungsgemeinschaft (DFG) through Grant SFB 951 “HIOS” is gratefully acknowledged. The work of D.A.E. has been partly supported by a DOC fellowship of the Austrian Academy of Sciences. The computational studies presented have been performed using

ASSOCIATED CONTENT

S Supporting Information *

Further details on the computational methodology, additional information regarding system setup and geometry optimizaE

dx.doi.org/10.1021/jp5084955 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

(24) Bumm, L. A.; Arnold, J. J.; Cygan, M. T.; Dunbar, T. D.; Burgin, T. P.; Jones, L.; Allara, D. L.; Tour, J. M.; Weiss, P. S. Are Single Molecular Wires Conducting? Science 1996, 271, 1705−1707. (25) Egger, D. A.; Rissner, F.; Rangger, G. M.; Hofmann, O. T.; Wittwer, L.; Heimel, G.; Zojer, E. Self-Assembled Monolayers of Polar Molecules on Au (111) Surfaces: Distributing the Dipoles. Phys. Chem. Chem. Phys. 2010, 12, 4291−4294. (26) Deutsch, D.; Natan, A.; Shapira, Y.; Kronik, L. Electrostatic Properties of Adsorbed Polar Molecules: Opposite Behavior of a Single Molecule and a Molecular Monolayer. J. Am. Chem. Soc. 2007, 129, 2989−2997. (27) Otálvaro, D.; Veening, T.; Brocks, G. Self-Assembled Monolayer Induced Au (111) and Ag (111) Reconstructions: Work Functions and Interface Dipole Formation. J. Phys. Chem. C 2012, 116, 7826− 7837. (28) Diez-Perez, I.; Hihath, J.; Hines, T.; Wang, Z. S.; Zhou, G.; Müllen, K.; Tao, N. Controlling Single-Molecule Conductance through Lateral Coupling of π Orbitals. Nat. Nanotechnol. 2011, 6, 226−231. (29) Sato, N.; Seki, K.; Inokuchi, H. Polarization Energies of Organic Solids Determined by Ultraviolet Photoelectron Spectroscopy. J. Chem. Soc., Faraday Trans. 2 1981, 77, 1621. (30) Neaton, J.; Hybertsen, M.; Louie, S. Renormalization of Molecular Electronic Levels at Metal-Molecule Interfaces. Phys. Rev. Lett. 2006, 97, 216405. (31) Refaely-Abramson, S.; Sharifzadeh, S.; Jain, M.; Baer, R.; Neaton, J. B.; Kronik, L. Gap Renormalization of Molecular Crystals from Density-Functional Theory. Phys. Rev. B 2013, 88, 081204. (32) Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for ab Initio Total-Energy Calculations Using a Plane-Wave Basis Set. Phys. Rev. B 1996, 54, 11169. (33) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865. (34) Soler, J. M.; Artacho, E.; Gale, J. D.; García, A.; Junquera, J.; Ordejón, P.; Sánchez-Portal, D. The SIESTA Method for ab Initio Order-N Materials Simulation. J. Phys.: Condens. Matter. 2002, 14, 2745. (35) Sancho, M. P. L.; Sancho, J. M.; Sancho, J. M. L.; Rubio, J. Highly Convergent Schemes for the Calculation of Bulk and Surface Greens Functions. J. Phys. F: Met. Phys. 1985, 15, 851. (36) Büttiker, M.; Imry, Y.; Landauer, R.; Pinhas, S. Generalized Many-Channel Conductance Formula with Application to Small Rings. Phys. Rev. B 1985, 31, 6207. (37) Xue, Y.; Datta, S.; Ratner, M. A. First-Principles Based Matrix Green’s Function Approach to Molecular Electronic Devices: General Formalism. Chem. Phys. 2002, 281, 151−170. (38) Darancet, P.; Widawsky, J. R.; Choi, H. J.; Venkataraman, L.; Neaton, J. B. Quantitative Current−Voltage Characteristics in Molecular Junctions from First Principles. Nano Lett. 2012, 12, 6250−6254. (39) Stukowski, A. Visualization and Analysis of Atomistic Simulation Data with OVITOThe Open Visualization Tool. Modell. Simul. Mater. Sci. Eng. 2010, 18, 015012. (40) Kokalj, A. XCrySDenA New Program for Displaying Crystalline Structures and Electron Densities. J. Mol. Graphics Modell. 1999, 17, 176−179. (41) Ramachandran, P.; Varoquaux, G. Mayavi: 3D Visualization of Scientific Data. Comput. Sci. Eng. 2011, 13, 40−51. (42) Paulsson, M.; Datta, S. Thermoelectric Effect in Molecular Electronics. Phys. Rev. B 2003, 67, 241403. (43) Reddy, P.; Jang, S. Y.; Segalman, R. A.; Majumdar, A. Thermoelectricity in Molecular Junctions. Science 2007, 315, 1568− 1571. (44) Widawsky, J. R.; Darancet, P.; Neaton, J. B.; Venkataraman, L. Simultaneous Determination of Conductance and Thermopower of Single Molecule Junctions. Nano Lett. 2011, 12, 354−358. (45) Romaner, L.; Heimel, G.; Zojer, E. Electronic Structure of Thiol-Bonded Self-Assembled Monolayers: Impact of Coverage. Phys. Rev. B 2008, 77, 045113.

the clusters of the division for high-performance computing at the Graz University of Technology.



REFERENCES

(1) Cuevas, J. C.; Scheer, E. Molecular Electronics−An Introduction to Theory and Experiment; World Scientific Publishing: Singapore, 2010. (2) Song, H.; Reed, M. A.; Lee, T. Single Molecule Electronic Devices. Adv. Mater. 2011, 23, 1583−1608. (3) McCreery, R. L.; Yan, H.; Bergren, A. J. A Critical Perspective on Molecular Electronic Junctions: There Is Plenty of Room in the Middle. Phys. Chem. Chem. Phys. 2013, 15, 1065−1081. (4) Akkerman, H. B.; Blom, P. W.; De Leeuw, D. M.; De Boer, B. Towards Molecular Electronics with Large-Area Molecular Junctions. Nature 2006, 441, 69−72. (5) Salomon, A.; Cahen, D.; Lindsay, S.; Tomfohr, J.; Engelkes, V. B.; Frisbie, C. D. Comparison of Electronic Transport Measurements on Organic Molecules. Adv. Mater. 2003, 15, 1881−1890. (6) Kushmerick, J. G.; Naciri, J.; Yang, J. C.; Shashidhar, R. Conductance Scaling of Molecular Wires in Parallel. Nano Lett. 2003, 3, 897−900. (7) Selzer, Y.; Cai, L.; Cabassi, M. A.; Yao, Y.; Tour, J. M.; Mayer, T. S.; Allara, D. L. Effect of Local Environment on Molecular Conduction: Isolated Molecule versus Self-Assembled Monolayer. Nano Lett. 2005, 5, 61−65. (8) Yaliraki, S. N.; Ratner, M. A. Molecule-Interface Coupling Effects on Electronic Transport in Molecular Wires. J. Chem. Phys. 1998, 109, 5036−5043. (9) Magoga, M.; Joachim, C. Conductance of Molecular Wires Connected or Bonded in Parallel. Phys. Rev. B 1999, 59, 16011. (10) Lang, N. D.; Avouris, P. Electrical Conductance of Parallel Atomic Wires. Phys. Rev. B 2000, 62, 7325. (11) Liu, R. U. I.; Ke, S. H.; Baranger, H. U.; Yang, W. Intermolecular Effect in Molecular Electronics. J. Chem. Phys. 2005, 122, 044703. (12) Landau, A.; Kronik, L.; Nitzan, A. Cooperative Effects in Molecular Conduction. J. Comput. Theor. Nanosci. 2008, 5, 535−544. (13) Landau, A.; Nitzan, A.; Kronik, L. Cooperative Effects in Molecular Conduction II: The Semiconductor−Metal Molecular Junction. J. Phys. Chem. A 2009, 113, 7451−7460. (14) Reuter, M. G.; Solomon, G. C.; Hansen, T.; Seideman, T.; Ratner, M. A. Understanding and Controlling Crosstalk between Parallel Molecular Wires. J. Phys. Chem. Lett. 2011, 2, 1667−1671. (15) Reuter, M. G.; Seideman, T.; Ratner, M. A. Molecular Conduction through Adlayers: Cooperative Effects Can Help or Hamper Electron Transport. Nano Lett. 2011, 11, 4693−4696. (16) Galperin, M.; Nitzan, A. Cooperative Effects in Inelastic Tunneling. J. Phys. Chem. B 2013, 117, 4449−4453. (17) Dubi, Y. Transport through Self-Assembled Monolayer Molecular Junctions: Role of In-Plane Dephasing. J. Phys. Chem. C 2014, DOI: 10.1021/jp503887p. (18) Vazquez, H.; Skouta, R.; Schneebeli, S.; Kamenetska, M.; Breslow, R.; Venkataraman, L.; Hybertsen, M. S. Probing the Conductance Superposition Law in Single-Molecule Circuits with Parallel Paths. Nat. Nanotechnol. 2012, 7, 663−667. (19) Natan, A.; Kronik, L.; Haick, H.; Tung, R. T. Electrostatic Properties of Ideal and Non-Ideal Polar Organic Monolayers: Implications for Electronic Devices. Adv. Mater. 2007, 19, 4103−4117. (20) Heimel, G.; Salzmann, I.; Duhm, S.; Koch, N. Design of Organic Semiconductors from Molecular Electrostatics. Chem. Mater. 2010, 23, 359−377. (21) Blumenfeld, M. L.; Steele, M. P.; Monti, O. L. Near-and FarField Effects on Molecular Energy Level Alignment at an Organic/ Electrode Interface. J. Phys. Chem. Lett. 2009, 1, 145−148. (22) Heimel, G.; Rissner, F.; Zojer, E. Modeling the Electronic Properties of π-Conjugated Self-Assembled Monolayers. Adv. Mater. 2010, 22, 2494−2513. (23) Egger, D. A.; Rissner, F.; Zojer, E.; Heimel, G. Polarity Switching of Charge Transport and Thermoelectricity in SelfAssembled Monolayer Devices. Adv. Mater. 2012, 24, 4403−4407. F

dx.doi.org/10.1021/jp5084955 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

(46) Natan, A.; Zidon, Y.; Shapira, Y.; Kronik, L. Cooperative Effects and Dipole Formation at Semiconductor and Self-AssembledMonolayer Interfaces. Phys. Rev. B 2006, 73, 193310. (47) Cornil, D.; Olivier, Y.; Geskin, V.; Cornil, J. Depolarization Effects in Self-Assembled Monolayers: A Quantum-Chemical Insight. Adv. Funct. Mater. 2007, 17, 1143. (48) Monti, O. L. A. Understanding Interfacial Electronic Structure and Charge Transfer: An Electrostatic Perspective. J. Phys. Chem. Lett. 2012, 3, 2342−2351.

G

dx.doi.org/10.1021/jp5084955 | J. Phys. Chem. C XXXX, XXX, XXX−XXX